The overall goal of this investigation is to ascertain the relationship between the rates and patterns of antibiotic use and the frequency and distribution of resistant bacteria, and to develop antibiotic use protocols to control the ascent and spread of drug resistant pathogens in individual patients and in communities of treated and untreated humans. Towards these ends, we will develop and analyze the properties of mathematical models of the population biology, epidemiology and (co)evolution R-plasmids in populations of bacteria in the environment and in infected mammals. To estimate the parameters of these models and test the validity of their assumptions and predictions, we will do in vitro and in vivo (laboratory mouse) experiments with R-plasmids in population of E. coli and other Gram negative bacteria. There are five specific aims: (1) To develop and analyze mathematical models of R-plasmid mediated drug resistance in populations of commensal and pathogenic bacteria in human communities subjected to different rates and patterns of antibiotic use. (2) To estimate, in vitro, the primary biological parameters of these models; the fitness burden associated with the carriage of novel R- plasmids and their rates loss by vegetative segregation in the absence of antibiotic-mediated selection. These estimates will be obtained for an array of conjugative R-plasmids in different strains and species of Gram negative bacteria, (3) To ascertain the nature, rate and consequences of (co)evolution on the fitness costs associated with the carriage of novel plasmids in E. coli and the rates these elements are lost by vegetative segregation. (4) To ascertain the effects of R-plasmid carriage on the fitness, dissemination, and virulence of bacteria infection laboratory mice. (5) To ascertain, the effects of temporal variation in antibiotic concentration on the ascent, dissemination and evolution of resistance genes in R-plasmids in bacteria infecting laboratory mice treated with single and multiple antibiotics.